When enough is enough: The worth of monitoring data in aquifer remediation design

Abstract

Given the high cost of data collection at groundwater contamination remediation sites, it is becoming increasingly important to make data collection as costeffective as possible. A Bayesian data worth framework is developed in an attempt to carry out this task for remediation programs in which a groundwater contaminant plume must be located and then hydraulically contained. The framework is applied to a hypothetical contamination problem where uncertainty in plume location and extent are caused by uncertainty in source location, source loading time, and aquifer heterogeneity. The goal is to find the optimum number and the best locations for a sequence of observation wells that minimize the expected cost of remediation plus sampling. Simplifying assumptions include steady state heads, advective transport, simple retardation, and remediation costs as a linear function of discharge rate. In the case here, an average of six observation wells was needed. Results indicate that this optimum number was particularly sensitive to the mean hydraulic conductivity. The optimum number was also sensitive to the variance of the hydraulic conductivity, annual discount rate, operating cost, and sample unit cost. It was relatively insensitive to the correlation length of hydraulic conductivity. For the case here, points of greatest uncertainty in plume presence were on average poor candidates for sample locations, and randomly located samples were not cost-effective. 1Now at Oak Ridge National Laboratory, Oak Ridge, Tennessee.Copyright !994 by the American Geophysical Union.Paper number 94WR01972. 0043-1397/94/94 WR-01972505.00 in reducing remediation costs. Second, collection of data should cease when its cost of acquisition is greater than its benefit in reducing remediation cost.We develop a Bayesian data worth framework for the case of aquifer remediation in which the location and extent of a contaminant plume are both uncertain, yet the plume must be contained hydraulically by pumping. Consider the hypothetical contamination problem displayed in Figures 1 and 2. Here, there exists a single plume which must be prevented from reaching a compliance surface. Although we display a plume in Figures 1 and 2, there is uncertainty about its location and extent because of three major factors: (1) the source location is only vaguely known, (2) the time that the source has been active is uncertain, and (3) the spatial variability of the hydraulic conductivity field is unknown. Given this uncertainty, in order to prevent with certainty the plume from reaching the compliance surface, a width much larger than the actual plume width must be contained. This results in excessively high remediation costs. Fortunately, uncertainty can be reduced by searching for the plume by installing monitoring wells and collecting water quality samples. The worth of installing a monitoring well is linked to how much the information that it provides can be used to reduce the width of the containment zone and hence the expected remediation cost. The foundation o...

Probabilistic human health risk assessment of degradation‐related chemical mixtures in heterogeneous aquifers: Risk statistics, hot spots, and preferential channels

“…The formation of these fast flow channels is typically associated with the presence of well-connected, highly permeable geological bodies or structures that can concentrate flow and solute transport [e.g., Knudby and Carrera, 2005; Incorporating hydrogeological uncertainty in human health predictions has been a topic of intense research in the past [e.g., Andričević and Cvetković, 1996;de Barros and Rubin, 2008;Cvetković and Molin, 2012;Rodak and Silliman, 2011;Andričević et al, 2012;Siirila and Maxwell, 2012;Atchley et al, 2013;de Barros and Fiori, 2014]. Probabilistic risk models allow one to determine the likelihood of risk exceeding a given regulatory threshold value [Tartakovsky, 2007], to delineate the spatial distribution of a plume for monitoring adaptation or intensification [James and Gorelick, 1994;Smalley et al, 2000;Maxwell et al, 2007;Fernandez-Garcia et al, 2012] and to better allocate characterization efforts to reduce the overall uncertainty of a given environmental performance metric [e.g., de Barros et al, 2009]. Most of the studies related to probabilistic risk analysis focused on the evaluation of human health risk posed by a single toxic compound.…”

Value of Information Analysis Project Gnome Site, New Mexico

“…The analysis framework is similar to that of other VOIAs related to environmental risk (James and Freeze, 1993;James and Gorelick, 1994;Yokota and Thompson, 2004;Feyen and Gorelick, 2005) but with minor modifications to suit the specified decision needs of this project. Unlike the analysis of James and Freeze (1993), this analysis does not address how to make a decision on various design alternatives.…”

“…It represents the maximum improvement that can be achieved by reducing the uncertainty to zero. In other words, it represents an estimate of the maximum budget that should be spent on exploration (James and Gorelick, 1994). The value of perfect information (VPI) is calculated as the difference between the prior project costs and the costs associated with knowing the exact extent of the radionuclide plume.…”

A hypothesis‐driven approach to optimize field campaigns

“…It can be used to optimize (1) what types of data (e.g., material parameters, state variables) to collect, (2) where to sample (e.g., the spatial layout and time schedule of observation networks), and (3) how to best excite the system to observe an informative response (e.g., designing tracer injections or hydraulic tests). Many applications in groundwater hydrology can be found in the literature [e.g., James and Gorelick, 1994;Reed et al, 2000a;Herrera and Pinder, 2005;Nowak et al, 2010;Leube et al, 2012].…”

“…However, data must be collected in a rational and goaloriented manner because field campaigns and laboratory analysis are expensive while budgets are limited [e.g., James and Gorelick, 1994].…”